KR-20260067792-A - Voice recognition and dialogue generation system considering the noise environment of construction sites

KR20260067792AKR 20260067792 AKR20260067792 AKR 20260067792AKR-20260067792-A

Abstract

The present invention relates to a voice recognition and dialogue generation system that takes into account the noise environment of a construction site. More specifically, it relates to a voice recognition and dialogue generation system that takes into account the noise environment of a construction site, which implements highly reliable voice recognition in a high-noise environment such as a construction site, and efficiently supports on-site communication by analyzing the worker's intent based on this and automatically generating dialogue. The present invention relates to a voice recognition and dialogue generation system considering the noise environment of a construction site, and may include a voice recognition unit that recognizes a voice signal of a worker at a construction site, an intent analysis unit that analyzes the worker's intent using an ontology knowledge structure from the voice signal recognized by the voice recognition unit and generates analysis data, and a dialogue generation unit that automatically generates feedback or a response from the analysis data generated by the intent analysis unit and provides it to the worker.

Inventors

김남현

Assignees

주식회사 엘젠

Dates

Publication Date: 20260513
Application Date: 20241106

Claims (5)

In a speech recognition and dialogue generation system considering the noise environment of a construction site, A voice recognition unit that recognizes a worker's voice signal at a construction site, An intention analysis unit that generates analysis data by analyzing the operator's intention using an ontology knowledge structure from a voice signal recognized by the above-mentioned voice recognition unit, and A speech recognition and dialogue generation system considering the noise environment of a construction site, comprising a dialogue generation unit that automatically generates feedback or a response from analysis data generated by the above-mentioned intention analysis unit and provides it to a worker.
In Article 1, The above voice recognition unit is, A beamforming microphone module that tracks the location of a worker's voice signal generation in real time through a multiple microphone array and a triangulation method, and selectively recognizes only the worker's voice signal, A background noise removal module that minimizes background noise by detecting and canceling background noise in real time, rather than the worker's voice signal, A frequency filtering module that filters out low- and high-frequency background noise generated at construction sites and maximizes voice clarity by emphasizing the worker's voice signal in the mid-frequency band, and A speech recognition and dialogue generation system considering the noise environment of a construction site, comprising an adaptive signal processing module that detects background noise fluctuating in real time and automatically adjusts the filtering level.
In Article 1, The above-mentioned intent analysis unit is a speech recognition and dialogue generation system that considers the noise environment of a construction site, which analyzes the worker's commands by mapping the worker's voice signals to work scenarios, equipment information, work procedures, and safety management elements of the construction site based on an ontology knowledge structure.
In Paragraph 3, If the above intent analysis unit analyzes the worker's voice signal as a worker's scream or emergency command, A voice recognition and dialogue generation system that takes into account the noise environment of a construction site, wherein the above dialogue generation unit automatically generates an emergency evacuation guidance dialogue or an emergency action dialogue, and provides the worker with the location of the worker voice signal generated by the above voice recognition unit.
In Article 1, Further including an equipment monitoring unit that continuously monitors and controls the status of the equipment according to the analysis data analyzed by the above-mentioned intention analysis unit, A voice recognition and dialogue generation system that takes into account the noise environment of a construction site, wherein the dialogue generation unit automatically generates feedback or a response from analysis data analyzed by the intent analysis unit and the status of equipment monitored and controlled by the equipment monitoring unit and provides it to the operator.

Description

Voice recognition and dialogue generation system considering the noise environment of construction sites The present invention relates to a voice recognition and dialogue generation system that takes into account the noise environment of a construction site. More specifically, it relates to a voice recognition and dialogue generation system that takes into account the noise environment of a construction site, which implements highly reliable voice recognition in a high-noise environment such as a construction site, and efficiently supports on-site communication by analyzing the worker's intent based on this and automatically generating dialogue. Since existing speech recognition technology was primarily developed under the assumption of quiet indoor environments, there is a problem in that it is difficult to guarantee accurate recognition rates in noisy environments such as construction sites. In other words, at construction sites, various types of noise—such as equipment operation, tool collisions, and people's conversations—occur simultaneously, making it difficult for speech recognition technology to clearly separate and recognize the worker's voice signal. Furthermore, in such noisy environments, the accuracy of intent analysis decreases due to the degradation of voice data quality, frequently resulting in the system failing to accurately understand the operator's commands or requests. For example, when an operator issues a command to stop equipment in a noisy environment, existing technology may fail to properly understand the intent of the command or execute an incorrect command. Furthermore, existing technologies are often limited to simple speech recognition and fail to efficiently support subsequent tasks such as analyzing the operator's intent or automatically generating dialogue. As a result, real-time communication between workers and the system is not smooth in complex and noisy environments such as construction sites, which can have a negative impact on work efficiency and safety. FIG. 1 is a configuration diagram of a voice recognition and dialogue generation system considering the noise environment of a construction site according to the present invention. FIG. 2 is a configuration diagram of the voice recognition unit of a voice recognition and dialogue generation system considering the noise environment of a construction site according to the present invention. The embodiments of the present disclosure are illustrative for the purpose of explaining the technical concept of the present disclosure. The scope of rights according to the present disclosure is not limited to the embodiments presented below or the specific description thereof. All technical and scientific terms used in this disclosure, unless otherwise defined, have the meaning generally understood by those skilled in the art to which this disclosure pertains. All terms used in this disclosure are selected for the purpose of further clarifying this disclosure and are not selected to limit the scope of the rights under this disclosure. Expressions such as “comprising,” “comprising,” “having,” etc. used in this disclosure should be understood as open-ended terms implying the possibility of including other embodiments, unless otherwise stated in the phrase or sentence containing such expressions. Unless otherwise stated, singular expressions described in this disclosure may include a plural meaning, and this applies likewise to singular expressions described in the claims. Hereinafter, a preferred embodiment of a voice recognition and dialogue generation system considering the noise environment of a construction site according to the present invention will be described in detail with reference to the attached drawings. FIG. 1 is a configuration diagram of a voice recognition and dialogue generation system considering the noise environment of a construction site according to the present invention, and FIG. 2 is a configuration diagram of a voice recognition unit of a voice recognition and dialogue generation system considering the noise environment of a construction site according to the present invention. The present invention relates to a voice recognition and conversation generation system that considers the noise environment of a construction site and efficiently supports on-site communication by implementing highly reliable voice recognition in a high-noise environment such as a construction site and, based on this, analyzing the worker's intention and automatically generating a conversation. The voice recognition and conversation generation system (100, hereinafter referred to as the 'system') that considers the noise environment of a construction site according to the present invention may include, as illustrated in FIGS. 1 and 2, a voice recognition unit (110) that recognizes a voice signal of a worker at a construction site, an intention analysis unit (120) that analyzes the worker's intention using an ontology knowledge str